Pub Date : 2021-10-01DOI: 10.4018/ijcini.20211001.oa4
A. Kharrat, M. Neji
We consider the problem of fully automatic brain tumor segmentation in MR images containing glioblastomas. We propose a three Dimensional Convolutional Neural Network (3D MedImg-CNN) approach which achieves high performance while being extremely efficient, a balance that existing methods have struggled to achieve. Our 3D MedImg-CNN is formed directly on the raw image modalities and thus learn a characteristic representation directly from the data. We propose a new cascaded architecture with two pathways that each model normal details in tumors. Fully exploiting the convolutional nature of our model also allows us to segment a complete cerebral image in one minute. The performance of the proposed 3D MedImg-CNN with CNN segmentation method is computed using dice similarity coefficient (DSC). In experiments on the 2013, 2015 and 2017 BraTS challenges datasets; we unveil that our approach is among the most powerful methods in the literature, while also being very effective.
{"title":"Segmentation of Brain Tumors Using Three-Dimensional Convolutional Neural Network on MRI Images 3D MedImg-CNN","authors":"A. Kharrat, M. Neji","doi":"10.4018/ijcini.20211001.oa4","DOIUrl":"https://doi.org/10.4018/ijcini.20211001.oa4","url":null,"abstract":"We consider the problem of fully automatic brain tumor segmentation in MR images containing glioblastomas. We propose a three Dimensional Convolutional Neural Network (3D MedImg-CNN) approach which achieves high performance while being extremely efficient, a balance that existing methods have struggled to achieve. Our 3D MedImg-CNN is formed directly on the raw image modalities and thus learn a characteristic representation directly from the data. We propose a new cascaded architecture with two pathways that each model normal details in tumors. Fully exploiting the convolutional nature of our model also allows us to segment a complete cerebral image in one minute. The performance of the proposed 3D MedImg-CNN with CNN segmentation method is computed using dice similarity coefficient (DSC). In experiments on the 2013, 2015 and 2017 BraTS challenges datasets; we unveil that our approach is among the most powerful methods in the literature, while also being very effective.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"47 1","pages":"1-17"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85491377","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/IJCINI.20211001.OA10
Hui Wang, Tie Cai, Wei-fang Cao
{"title":"Citrus Huanglongbing Recognition Algorithm Based on CKMOPSO","authors":"Hui Wang, Tie Cai, Wei-fang Cao","doi":"10.4018/IJCINI.20211001.OA10","DOIUrl":"https://doi.org/10.4018/IJCINI.20211001.OA10","url":null,"abstract":"","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"26 1","pages":"1-11"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91029984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/IJCINI.20211001.OA38
T. Gururaj, G. Siddesh
{"title":"Hybrid Approach for Enhancing Performance of Genomic Data for Stream Matching","authors":"T. Gururaj, G. Siddesh","doi":"10.4018/IJCINI.20211001.OA38","DOIUrl":"https://doi.org/10.4018/IJCINI.20211001.OA38","url":null,"abstract":"","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"28 1","pages":"1-18"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74031161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/ijcini.20211001.oa9
Muhammad Salman Khan, Rene Richard, Heather Molyneaux, Danick Cote-Martel, Henry Jackson Kamalanathan Elango, Steve Livingstone, Manon Gaudet, David V. Trask
Security and Information Event Management (SIEM) systems require significant manual input; SIEM tools with machine learning minimizes this effort but are reactive and only effective if known attack patterns are captured by the configured rules and queries. Cyber threat hunting, a proactive method of detecting cyber threats without necessarily knowing the rules or pre-defined knowledge of threats, still requires significant manual effort and is largely missing the required machine intelligence to deploy autonomous analysis. This paper proposes a novel and interactive cognitive and predictive threat-hunting prototype tool to minimize manual configuration tasks by using machine intelligence and autonomous analytical capabilities. This tool adds proactive threat-hunting capabilities by extracting unique network communication behaviors from multiple endpoints autonomously while also providing an interactive UI with minimal configuration requirements and various cognitive visualization techniques to help cyber experts quickly spot events of cyber significance from high-dimensional data.
{"title":"Cyber Threat Hunting: A Cognitive Endpoint Behavior Analytic System","authors":"Muhammad Salman Khan, Rene Richard, Heather Molyneaux, Danick Cote-Martel, Henry Jackson Kamalanathan Elango, Steve Livingstone, Manon Gaudet, David V. Trask","doi":"10.4018/ijcini.20211001.oa9","DOIUrl":"https://doi.org/10.4018/ijcini.20211001.oa9","url":null,"abstract":"Security and Information Event Management (SIEM) systems require significant manual input; SIEM tools with machine learning minimizes this effort but are reactive and only effective if known attack patterns are captured by the configured rules and queries. Cyber threat hunting, a proactive method of detecting cyber threats without necessarily knowing the rules or pre-defined knowledge of threats, still requires significant manual effort and is largely missing the required machine intelligence to deploy autonomous analysis. This paper proposes a novel and interactive cognitive and predictive threat-hunting prototype tool to minimize manual configuration tasks by using machine intelligence and autonomous analytical capabilities. This tool adds proactive threat-hunting capabilities by extracting unique network communication behaviors from multiple endpoints autonomously while also providing an interactive UI with minimal configuration requirements and various cognitive visualization techniques to help cyber experts quickly spot events of cyber significance from high-dimensional data.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"11 I 1","pages":"1-23"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87314829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/ijcini.20211001.oa8
Fei Zhao, Qinghui Xu, Sanyou Zeng
Antenna design often requires dealing with multiple constraints in the requirements, and the designs can be modeled as constrained optimization problems (COPs). However, the constraints are usually very strange, and then the feasible solutions are hard to be found. At the same time, the robustness for antenna design is an important consideration as well. To solve the above issues, the combination of differential evolution algorithm (DE) and 3D-printing technique is presented to design a new crooked-wire antenna. In the design process, DE is adopted to handle the constraints since DE is simple and efficient in finding feasible solutions. The objective of the modeled COP, which is the sum of variance of the gain, axial ratio, and VSWR over the frequency band, is used to enhance the robustness of the antenna and widen the frequency band without additional computational cost. The precision of fabricating the antenna is ensured by using 3D-printing. The design of the NASA LADEE satellite antenna is chosen as an example to verify the method of this paper.
{"title":"Design of a Crooked-Wire Antenna by Differential Evolution and 3D Printing","authors":"Fei Zhao, Qinghui Xu, Sanyou Zeng","doi":"10.4018/ijcini.20211001.oa8","DOIUrl":"https://doi.org/10.4018/ijcini.20211001.oa8","url":null,"abstract":"Antenna design often requires dealing with multiple constraints in the requirements, and the designs can be modeled as constrained optimization problems (COPs). However, the constraints are usually very strange, and then the feasible solutions are hard to be found. At the same time, the robustness for antenna design is an important consideration as well. To solve the above issues, the combination of differential evolution algorithm (DE) and 3D-printing technique is presented to design a new crooked-wire antenna. In the design process, DE is adopted to handle the constraints since DE is simple and efficient in finding feasible solutions. The objective of the modeled COP, which is the sum of variance of the gain, axial ratio, and VSWR over the frequency band, is used to enhance the robustness of the antenna and widen the frequency band without additional computational cost. The precision of fabricating the antenna is ensured by using 3D-printing. The design of the NASA LADEE satellite antenna is chosen as an example to verify the method of this paper.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"425 1","pages":"1-16"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84946308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/IJCINI.20211001.OA23
Ting Wang, Wing W. Y. Ng, Wendi Li, S. Kwong
Activation functions such as tanh and sigmoid functions are widely used in deep neural networks (DNNs) and pattern classification problems. To take advantage of different activation functions, this work proposes the broad autoencoder features (BAF). The BAF consists of four parallel-connected stacked autoencoders (SAEs), and each of them uses a different activation function, including sigmoid, tanh, relu, and softplus. The final learned features can merge by various nonlinear mappings from original input features with such a broad setting. It not only helps to excavate more information from the original input features through utilizing different activation functions, but also provides information diversity and increases the number of input nodes for classifier by parallel-connected strategy. Experimental results show that the BAF yields better-learned features and classification performances.
{"title":"Broad Autoencoder Features Learning for Classification Problem","authors":"Ting Wang, Wing W. Y. Ng, Wendi Li, S. Kwong","doi":"10.4018/IJCINI.20211001.OA23","DOIUrl":"https://doi.org/10.4018/IJCINI.20211001.OA23","url":null,"abstract":"Activation functions such as tanh and sigmoid functions are widely used in deep neural networks (DNNs) and pattern classification problems. To take advantage of different activation functions, this work proposes the broad autoencoder features (BAF). The BAF consists of four parallel-connected stacked autoencoders (SAEs), and each of them uses a different activation function, including sigmoid, tanh, relu, and softplus. The final learned features can merge by various nonlinear mappings from original input features with such a broad setting. It not only helps to excavate more information from the original input features through utilizing different activation functions, but also provides information diversity and increases the number of input nodes for classifier by parallel-connected strategy. Experimental results show that the BAF yields better-learned features and classification performances.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"6 1","pages":"1-15"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84794305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/IJCINI.20211001.OA43
Yanping Yang, Ruiguang Li
For the system with unknown statistical property noises, the property that the energies of the system noise and the observation noise are limited is utilized in this paper. On this basis, two novel fusion algorithms are proposed for ship-integrated navigation with the relative navigation information broadcasted by the automatic identification systems (AISs) in the adjacent ships. Firstly, an H¥ fusion filtering algorithm is given to deal with the navigation observation messages under the centralized fusion framework. The integrated navigation method based on this algorithm cannot deal with the asynchronous navigation messages in real time. Therefore, a sequential H¥ fusion-filtering algorithm is also given to sequentially deal with the asynchronous navigation messages. Finally, a computer simulation is employed to illustrate the validity and feasibility of the sequential method.
{"title":"Relatively-Integrated Ship Navigation by H¥ Fusion Filters","authors":"Yanping Yang, Ruiguang Li","doi":"10.4018/IJCINI.20211001.OA43","DOIUrl":"https://doi.org/10.4018/IJCINI.20211001.OA43","url":null,"abstract":"For the system with unknown statistical property noises, the property that the energies of the system noise and the observation noise are limited is utilized in this paper. On this basis, two novel fusion algorithms are proposed for ship-integrated navigation with the relative navigation information broadcasted by the automatic identification systems (AISs) in the adjacent ships. Firstly, an H¥ fusion filtering algorithm is given to deal with the navigation observation messages under the centralized fusion framework. The integrated navigation method based on this algorithm cannot deal with the asynchronous navigation messages in real time. Therefore, a sequential H¥ fusion-filtering algorithm is also given to sequentially deal with the asynchronous navigation messages. Finally, a computer simulation is employed to illustrate the validity and feasibility of the sequential method.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"48 1","pages":"1-12"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86799821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/IJCINI.20211001.OA29
Weiwei Yu, Li Zhang, Chengwang Xie
Many-objective optimization problems (MaOPs) refer to those multi-objective problems (MOPs) with more than three objectives. In order to solve MaOPs, a multi-objective particle swarm optimization algorithm based on new fitness assignment and multi cooperation strategy (FAMSHMPSO) is proposed. Firstly, this paper proposes a new fitness allocation method based on fuzzy information theory to enhance the convergence of the algorithm. Then a new multi-criteria mutation strategy is introduced to disturb the population and improve the diversity of the algorithm. Finally, the external files are maintained by the three-point shortest path method, which improves the quality of the solution. The performance of FAMSHMPSO algorithm is evaluated by evaluating the mean value, standard deviation, and IGD+ index of the target value on dtlz test function set of different targets of FAMSHMPSO algorithm and other five representative multi-objective evolutionary algorithms. The experimental results show that FAMSHMPSO algorithm has obvious performance advantages in convergence, diversity, and robustness.
{"title":"Many-Objective Particle Swarm Optimization Algorithm Based on New Fitness Allocation and Multiple Cooperative Strategies","authors":"Weiwei Yu, Li Zhang, Chengwang Xie","doi":"10.4018/IJCINI.20211001.OA29","DOIUrl":"https://doi.org/10.4018/IJCINI.20211001.OA29","url":null,"abstract":"Many-objective optimization problems (MaOPs) refer to those multi-objective problems (MOPs) with more than three objectives. In order to solve MaOPs, a multi-objective particle swarm optimization algorithm based on new fitness assignment and multi cooperation strategy (FAMSHMPSO) is proposed. Firstly, this paper proposes a new fitness allocation method based on fuzzy information theory to enhance the convergence of the algorithm. Then a new multi-criteria mutation strategy is introduced to disturb the population and improve the diversity of the algorithm. Finally, the external files are maintained by the three-point shortest path method, which improves the quality of the solution. The performance of FAMSHMPSO algorithm is evaluated by evaluating the mean value, standard deviation, and IGD+ index of the target value on dtlz test function set of different targets of FAMSHMPSO algorithm and other five representative multi-objective evolutionary algorithms. The experimental results show that FAMSHMPSO algorithm has obvious performance advantages in convergence, diversity, and robustness.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"25 1","pages":"1-23"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75471118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/IJCINI.20211001.OA34
Anand Handa, Rashi Agarwal, Narendra Kohli
Due to the highly variant face geometry and appearances, facial expression recognition (FER) is still a challenging problem. CNN can characterize 2D signals. Therefore, for emotion recognition in a video, the authors propose a feature selection model in AlexNet architecture to extract and filter facial features automatically. Similarly, for emotion recognition in audio, the authors use a deep LSTM-RNN. Finally, they propose a probabilistic model for the fusion of audio and visual models using facial features and speech of a subject. The model combines all the extracted features and use them to train the linear SVM (support vector machine) classifiers. The proposed model outperforms the other existing models and achieves state-of-the-art performance for audio, visual, and fusion models. The model classifies the seven known facial expressions, namely anger, happy, surprise, fear, disgust, sad, and neutral, on the eNTERFACE’05 dataset with an overall accuracy of 76.61%.
{"title":"Audio-Visual Emotion Recognition System Using Multi-Modal Features","authors":"Anand Handa, Rashi Agarwal, Narendra Kohli","doi":"10.4018/IJCINI.20211001.OA34","DOIUrl":"https://doi.org/10.4018/IJCINI.20211001.OA34","url":null,"abstract":"Due to the highly variant face geometry and appearances, facial expression recognition (FER) is still a challenging problem. CNN can characterize 2D signals. Therefore, for emotion recognition in a video, the authors propose a feature selection model in AlexNet architecture to extract and filter facial features automatically. Similarly, for emotion recognition in audio, the authors use a deep LSTM-RNN. Finally, they propose a probabilistic model for the fusion of audio and visual models using facial features and speech of a subject. The model combines all the extracted features and use them to train the linear SVM (support vector machine) classifiers. The proposed model outperforms the other existing models and achieves state-of-the-art performance for audio, visual, and fusion models. The model classifies the seven known facial expressions, namely anger, happy, surprise, fear, disgust, sad, and neutral, on the eNTERFACE’05 dataset with an overall accuracy of 76.61%.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"40 1","pages":"1-14"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83156611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.4018/ijcini.20211001.oa27
Haixia Gui, Banglei Zhao, Huizong Li, Wanliu Che
Overlapping coalition formation is a very active research field in multi-agent systems (MAS). In overlapping coalition, each agent can participate in different coalitions corresponding to multiple tasks at the same time. As each agent has limited resources, resource conflicts will occur. In order to resolve resource conflicts, we develop an improved encoding revision algorithm in this paper which can revise an invalid two-dimensional binary encoding into a valid one by checking the encoding for each row. To verify the effectiveness of the algorithm, differential evolution was used as the experimental platform and compared with Zhang et al. The experimental results show that the algorithm in this paper is superior to Zhang et al. in both solution quality and encoding revision time.
{"title":"Application of an Encoding Revision Algorithm in Overlapping Coalition Formation","authors":"Haixia Gui, Banglei Zhao, Huizong Li, Wanliu Che","doi":"10.4018/ijcini.20211001.oa27","DOIUrl":"https://doi.org/10.4018/ijcini.20211001.oa27","url":null,"abstract":"Overlapping coalition formation is a very active research field in multi-agent systems (MAS). In overlapping coalition, each agent can participate in different coalitions corresponding to multiple tasks at the same time. As each agent has limited resources, resource conflicts will occur. In order to resolve resource conflicts, we develop an improved encoding revision algorithm in this paper which can revise an invalid two-dimensional binary encoding into a valid one by checking the encoding for each row. To verify the effectiveness of the algorithm, differential evolution was used as the experimental platform and compared with Zhang et al. The experimental results show that the algorithm in this paper is superior to Zhang et al. in both solution quality and encoding revision time.","PeriodicalId":43637,"journal":{"name":"International Journal of Cognitive Informatics and Natural Intelligence","volume":"27 1","pages":"1-20"},"PeriodicalIF":0.9,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81685971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：